From the past, systems have been proposed which simultaneously image a common subject using two right and left video cameras and simultaneously output the right and left videos to display a stereoscopic image. However, the system using two video cameras tends to be large in size and lacks mobility, deviation between optical axes of the right and left video cameras may occur, and it is difficult to obtain an image having a proper parallax. For example, there are such concerns that the distance between two video cameras may be increased, right and left optical axes may be deviated during zooming due to an individual difference between lenses of the video cameras, and the sizes of right and left screens may be different. Further, when a convergence operation of orienting right and left video cameras toward a subject is performed in a focus operation, there is a concern that the right and left video cameras may be deviated, for example, vertically.
When the deviation or the like occurs in the optical axes of the right and left video cameras, an optical system of a user who views a stereoscopic image is urged to carry out information processing different from normal processing, and thus visual fatigue may be caused. Further, when the user views overlapping right and left images without using stereoscopic glasses, the user may view a subject doubly, and thus an unnatural video may be created.
Accordingly, an imaging apparatus has been proposed in which light from a subject is separated into two light fluxes by a mirror to image the subject in a region of one lens pupil (for example, see PTL 1). In the imaging apparatus, right and left video data for a stereoscopic view can be obtained by performing imaging on each of the two separated light fluxes.